preferred direction of karst in the kuala ... direction of karst in the kuala lumpur limestone...

PREFERRED DIRECTION OF KARST IN THE

KUALA LUMPUR LIMESTONE FORMATION:

A SMART TUNNEL CASE STUDY

Hareyani Zabidi1 and Michael H. De Freitas

2

1School of Materials and Mineral Resources Engineering, Universiti Sains Malaysia, 14300 Nibong Tebal,

Penang, Malaysia 2Department of Civil and Environmental Engineering, Skempton Building, Imperial College, SW7 2AZ,

London, UK

Corresponding author:[email protected], phone: 045996124, fax: 045941011

Abstract

In this paper, a karstified limestone formation beneath the city of Kuala Lumpur is studied in

reference to its relationship between the geological structure and river patterns. A preliminary

prediction of karst that was likely to be readily developed was made based on the existing

geological map, where the basic geology of the area was expected to be related to river patterns;

which suspected can be obtained from the direction of fractures expected in the limestone. Two

directions appear to be special interest: N030° to 070° and N120° to 150°; the investigation was

then carried out in the field, using the two sites exposed for construction of the SMART tunnel

under Kuala Lumpur to verify the prediction made in the desk study analysis. The method used for

this study and the evidence are described.

Keywords: Fractures, Karst, Rivers and streams, Kuala Lumpur, Limestone

Introduction

Most of Kuala Lumpur, including its commercial centre, is founded on alluvium overlying

limestone of Silurian age, in which Extreme karst, classified as KV according to Waltham

& Fookes (2003), has been extensively developed. This buried karst landscape of Kuala

Lumpur can be further grouped into two classes in accordance with its dimensions and

characteristics. The first class is a Karst Scale 1 (K1) that is well developed, creating a

highly irregular level of rock below the alluvium, varying in elevation by tens of metres

over tens of metres and containing numerous voids, many of which have collapsed and are

partially filled. The second is a Karst Scale 2 (K2) that is much smaller and concentrated

on fractures having been investigated concerning an openness (measured in centimetres)

and freshness that is still being actively developed (Zabidi and deFreitas, 2010). This

smaller scale karst creates many problems for the design and construction of civil

engineering structures in Kuala Lumpur because the void of it is so permeable that it

provides major sources of inflow and locations for considerable grout loss, and zones of

great weakness under load. The highly irregular topography of limestone rock head in the

Kuala Lumpur region was first discovered in opencast tin mines, nearly 150 years ago.

Deep borehole records from construction sites in the city confirm its wide spread

occurrence (Tan, 1986a & 1986b). This karst feature is generally believed to have been

developed during the Quaternary, although it is possible that considerable dissolution also

has occurred prior to the deposition of the Permo-Carboniferous Kenny Hill Formation;

this paleo-landscape has been buried by alluvium to form the current landscape of Kuala

Lumpur (Chan and Hong, 1985).

The construction of high rise buildings on a flat alluvial plain that conceals the highly

irregular topography of limestone bedrock has always been a challenge for engineers in

Kuala Lumpur (Ibrahim and Fang, 1985; Tan and Komoo, 1990) and considerable

ASEAN Engineering Journal Part C, Vol 1 No 3, ISSN 2229-127X p.152

geotechnical problems have occurred during the construction of many engineering projects,

large and small, in the limestone formation. Well recorded case histories for the

construction of the Petronas Twin Tower (Tan, 1996; Pollalis, 2002), the Berjaya Times

Square Complex (Gue and Tan, 2001), the Pan Pacific Hotel (Mitchell, 1985) and the

SriMARA Complex (Tan et. al., 1985) illustrate the types of conditions encountered. The

difficulty in locating them is eliminated by ground investigations that depend heavily on

vertical boreholes to study the complexity of the ground; but, these give no guarantee of

finding either karst at this scale or trends in such karst development. Therefore, the other

methods of prediction are needed. In this paper studies of the relationship between

drainage patterns, fracture patterns and K2 karst forms are reported, and their use for

predicting karst during desk studies are assessed.

Geology of Kuala Lumpur

Kuala Lumpur lies on a flat alluvial plain within the broad valley of the Klang River,

bounded by high hills predominantly of granitic rock to west and east. The Klang River

drains a catchment area of approximately 1,288km2, traverses a distance of nearly 120km

while crossing two states, Selangor and Wilayah, before discharging into the Straits of

Melaka. The geology of the study area consists of sediments ranging in age from Middle

Ordovician to possibly Permian, and a granitic body intruded during the Late Triassic (Fig.

1). At the base of the sequence is the Hawthornden Formation (Middle Ordovician to

Middle Silurian), overlain by the Kuala Lumpur Limestone Formation (Middle Silurian to

Lower-Middle Devonian) (Gobbett, 1964). The Hawthornden Formation is a mixture of

quartz – mica amphibolites and carbonaceous schists, phyllites and quartzites. The

overlying Kuala Lumpur Limestone Formation is composed of fine to coarse grained,

white to grey, predominantly recrystallised limestones, with local developments of

dolomitic limestone and dolomites, all with few impurities. These Lower Palaeozoic

formations experienced their first phase of folding during the Devonian to form east-west

fold axes. An extensive period of uplift, weathering and erosion followed during which

karst was developed in the Kuala Lumpur Limestone Formation. These Lower Palaeozoic

formations are overlain unconformably by the shales, mudstones and sandstones of the

Kenny Hill Formation, which accumulated towards the end of the Carboniferous and the

start of the Permian. These sediments were folded by a second tectonic during the Late

Triassic. This activity strongly deformed the Lower Paleozoic rocks to produce the

metamorphic grades and folded strata to follow a north-south trend. The Lower Palaeozoic

sequence has bedding dips that are commonly steep and overturned; contrasting with the

more gentles dips of the Kenny Hill Formation overlying them. The country rock was then

intruded by granite, estimated to be either broadly contemporaneous or younger than the

second phase of folding, occurring towards the Late Triassic. The last period of

deformation is NE-SW and NW-SE trending faulting, which has affected all the formations,

including the granite. The fault zones of the Ampang Fault, trending at N285°, and the

Gombak Fault, at N200°, have markedly displaced the Kuala Lumpur Limestone

Formation and can be expected to have affected its hydraulic conductivity.

Method of Prediction

An assumption has been made that fractures offer lines of decreased hydraulic resistance to

groundwater flow and therefore they are more easily exploited by weathering and erosion

processes than their adjacent rock (Verstappen, 1960; Paton, 1964; Gobbett, 1965; Sower,

1975 and Ericson et al., 2004). With this assumption, the existing geological and

geomorphological data can be interpreted and potential links between the drainage pattern


and possible directions for karst can be identified. Thus, by studying the pattern of river

and structural orientation from the geological map, an initial prediction for the orientation

of karst can be made. Four steps were followed to explore this approach: (i) the structure of

the limestone was defined, using the 1: 63,600 geological map of the region, to obtain the

likely direction of joints; (ii) the drainage network of the region was studied using the same

scale map (in fact in this study it was the same map) to obtain the directions and

frequencies of linear river segments; (iii) the river directions from the drainage analysis

were then compared with the structural directions obtained from the fracture analysis,

which showed that rivers and structures shared certain directions but not other, suggesting

the needs to know whether the structural directions in which there are no rivers (i.e. no

water at the surface) are the same as those in which karst has most severely developed, and

if so, whether this is a means of predicting likely preferential directions of karst; and (iv)

the correctness of this approach was confirmed by field observations in Kuala Lumpur. In

this exercise, geological mapping was carried out at the two previously excavated boxes,

which was constructed as part of the Stormwater Management and Road Tunnel (SMART)

project.

Figure 1. A simplified version of the geological map of Selangor, Sheet 94 (Yin, 1967).

Fractures Analysis

A joint is a fissure in rock across which there is no appreciable displacement; it develops in

almost all types of rock and in diverse tectonic settings. Systematic joint sets provide

crucial information on the directions of principal stress involved in the formation of the

ground. In slightly deformed, horizontal or near-horizontally bedded rock, failure in

tension can occur, to form an extensional joint oriented parallel to the maximum stress, σ1

and intermediate stress, σ2 and perpendicular to the minimum stress, σ3 (Hatcher, 1995). In

folded rock, extensional joints, normally cut perpendicular to the bedding, can also form as

shear joints, with oblique orientations to the principle stresses (Price, 1966).

The predictions of preferential directions of karst started with an analysis of the

structural geology of the area based on the 1: 63,360 New Series geological map for the

Selangor area, Sheet 94 (Yin, 1967). The Kuala Lumpur Limestone Formation contains


many folds and the axes of a few recorded folds on the geological map as shown in Fig. 2.

From this the likely directions of fractures (joints and faults) were identified. Fracture

orientations were predicted from the orientation of fold axes and bedding planes, based on

the assumption that orientations of strike and dip can be used to deduce another three joints

orientations according to the fold – fracture relationship for obtaining extensional joint,

longitudinal joint and two shear joints (Price, 1966). That is possible by using the stereonet;

E1 is bedding reading from the geological map, whereas the other three readings are

collected from the stereonet (Fig. 2). The accumulation of fracture line directions was then

displayed, measured from Grid North (which is within 0° 3' East of Magnetic North in this

area), by using a rose diagram to produce percentages in any given direction grouped in

batches of 10° from North, with all the readings are grouped in 18 equal sizes, N001° –

N010°, etc. The 3D characterization of these fractures could equally be shown using a

stereonet (Fig. 3), however since most of these fractures are vertical or near vertical, simple

rose diagrams are used for clarity.

Figure 2. Fracture analysis carried out in the Kuala Lumpur Limestone Formation

Drainage Analysis

Using the same map, the stream directions in the area, which exhibit a trellis-like pattern

indicating some degree of structural control, were analysed. This orientation has frequently

been mentioned by the previous researchers as the prominent structure orientation, (e.g.

Stauffer (1968), Shu (1969), Tjia (1972), and Ong (1978)). Strong correlations between

fracture patterns and drainage lines have been long suspected by many researchers in the

past (Herold, et al., 2000 & Mayer, et al., 2003). An assumption was made at this stage in

the work that some fracture lines or zones within the host rock offer lines or zones of

decreased resistance to dissolution, and is more easily exploited by weathering and erosion

processes than more massive adjacent rock. The end product of the weathering and erosion

process is a stream or river channel, developed along a fracture orientation.

Stream analysis was carried out by tracing the drainage network shown on the 1:63,360

geological map for all geological formations in the study area and dividing the stream

channels into six different groups based on their direction from magnetic North to produce

a stream line map (Fig. 4). The lines were grouped into 6 classes of equal orientation size,

(N001°–N030°), (N031°–N060°), (N061°-N090°), etc. Figure 4 shows part of the study


area, located further Northwest of Kuala Lumpur area. Apparently, it is rather obvious that

the first upper part of the traced map is dominated with the NW direction of lines, marked

in red pen. A comparison was then completed between the orientations of the drainage

channels in the Kuala Lumpur Limestone Formation and the orientations of the drainage

channels in the other geological formations including the granites, the Kenny Hill

Formation and the Hawthornden Formation; i.e. formations that were older and younger

than the Kuala Lumpur Formation. The accumulated number of measured drainage

channels for each geological formation was then grouped into 18 classes of equal size,

(N001°–010°), (N011°–N020°), (N021°–N030°), etc., i.e., using the same intervals as used

for fractures, and plotted to make a complete diagram for the evidence from each outcrop

type (Fig. 5).

Figure 3. Joint orientations in the Kuala Lumpur Limestone Formation, as deduced from

the geological map. The two prominent orientations are trending from N030 – N070 and

N130 – N140. (N = number of readings) Radial division = 2.5%; outer circle = 30%

Analysis of the drainage patterns developed on the outcrops of each formation revealed

a strong link between their direction and the direction of underling geological structures;

(Fig. 5); in this study, the oldest rocks, and thus the ones that have undergone the most

phases of deformation, are on the top left (Hawthornden Formation) and the youngest

(granites) are on the bottom right. The stream directions in the Hawthornden Formation,

which is older than the Kuala Lumpur Limestone Formation, are more uniformly

distributed than those in the Kenny Hill Formation, which is younger than the Kuala

Lumpur Limestone Formation and where, for example, almost none are present between

(N360°-N040°); the granites have a drainage pattern largely independent of the fold history

of their country rock. Thus the stream patterns appear to reflect geological structure and

history. The same structures might be expected in the Kuala Lumpur Limestone, because it

occurs between the Hawthornden and Kenny Hill Formations, and to some extent the

structures do follow the pattern but there are significant differences; these are shown in Fig.

6. The figure shows the comparison of drainage patterns traced in the Kuala Lumpur

Limestone Formation and the other formations combined together in one rose diagram.


Figure 4. Drainage analysis carried out based on the geological map for the Selangor area,

showing part of the study area, located further Northwest of Kuala Lumpur area.

Figure 5. Rose diagrams of drainage direction for streams on all the geological formations

in the Kuala Lumpur area.

The stream patterns in the Kuala Lumpur Limestone Formation have a prominent trend

of N360°-N010°, N090°-N120° and N150°-N160°. The stream channels from the other

formations produce a fairly uniform distribution of orientations as shown in Fig. 6b with a

few slightly higher percentages, as in the trends of N070°-N080° and N120°-N130°. There

are no streams on the trend of N360°-N030°, but they are strongly developed between

N030° and N180°. The contrast between the drainage orientations obtained for the Kuala

Lumpur Limestone Formation and the other formations could be linked to an extensive

underground drainage system in the limestone formation; a condition that would explain

the lack of surface drainage across its mapped outcrop, as seen in the 1: 63,360 geological

map. The relationship of the orientations between fracture and drainage demonstrates that

there is a close and interesting link between them. That might be diagnostic of karst.

Karst Pattern Prediction

Prominent directions of drainage channels exist in the limestone between N000°-N010°,

N090°-N120°, and N150°-N160°. In contrast there are almost no streams in the directions

N030°-N090°. The structural information for fractures in the Kuala Lumpur Limestone


(Fig. 3) shows a prominent direction between N030°-N070° and another around N130°-

N140°. By comparing with Figs. 3 and 6, Figure 7 suggests that the lack of surface water

on the outcrop of the Kuala Lumpur Limestone Formation in the direction N030°-N090°

may be related to the structural direction N030°-N070° (Fig. 7a). This difference is seen

again when comparing it with the stream orientations for all the outcrops (Fig. 7b). These

differences in the dominant orientations between the Kuala Lumpur Limestone Formation

and the other formations could be linked to the existence of solution features below ground

level. Therefore, based on the analysis, karst in the Kuala Lumpur Limestone Formation

can be predicted to form most favourably, i.e. to have the greatest risk of karst occurring,

in the orientation N030°-N070° and N120°-N150°, following the fractures patterns in the

Kuala Lumpur Limestone Formation. Nonetheless, the prediction made from this

geological map study requires further validation and thus geological mapping at the two

previously exposed boxes constructed for the SMART project has been carried out.

Figure 6. Different between prominent directions of the drainage lines in the Kuala

Lumpur Limestone Formation and the other formations in the study area.

Figure 7. Fracture orientation in the Kuala Lumpur Limestone Formation, as deduced from

the geological map of Selangor.

Mapping of Karst at the SMART Project Sites

The geological data available to check these predictions came from detailed geological

mapping of the excavation sites located at the Kampung Pandan Roundabout (North

Junction Box) and at the Sungai Besi (South Junction Box) excavated for the Stormwater

Management and Road Tunnel (SMART) tunnel (SMART, 2005). At the North Junction


Box, exposure of karst features occurs on the excavated sides of a rectangular box,

approximately 500m in length, 20m in width and 30m in depth. The rock profile in the

Kampung Pandan Roundabout can generally be classified as strongly deformed and

heavily fractured (Bieniawski, 1989) and contains numerous karst features; e.g. between

Ch300 and Ch380, two large zones of depression, of the type previously described as Karst

Scale 1(K1) were exposed on both sides of the box (Fig. 8); their trend followed the

direction (N040°), and this is close to that of the Gombak Fault (N020°) direction. So it is

possible that features of K1 scale karst are related to major deformation episodes in the

Kuala Lumpur area. The excavation also revealed a clear evidence of left-lateral strike-slip

faulting directed along N330°, the trend of the Ampang Fault and the Klang Gates Quartz

Reef (Fig. 1). The highly broken limestone seen in Fig. 8 represents the (K2) scale karst;

joints are a dominant character of the rock mass and contain a wide range of cavity sizes.

The second site is the South Junction Box, which is 70 m long, 20 m wide and

approximately 30 m high, and located between the KL-Seremban Expressway and Jalan

Sungai Besi, at approximately at Ch7500 of the SMART tunnel. The limestone here is

generally of far better quality than that seen in the Kampung Pandan Roundabout box;

being massive, dense and frequently recrystallised, reflecting its proximity to the granite

(Fig 1). Joints are dominantly orientated N160° and N230°, and are not as frequent as seen

in the North Junction Box. The limestone has within it a bed of originally argillaceous

sediment, now phyllite, orientated at N020°, in which the weathering has taken place,

shown by the occurrence of a 30 cm thick brown yellow discolouration, but apart from that

the limestone is fresh and solid rock.

Figure 8. Two large zones of depression in the Kampung Pandan Roundabout, classified as

Karst Scale 1(K1) were exposed on both sides of the box.

Conclusion

In this study, the results obtained from geological maps and fracture analyses were

compared and further validated based on observations carried out at the two selected sites

in Kuala Lumpur. Based on the assumption that there could be a close relationship between

drainage, fracture pattern and karst formation, the geological map analysis was carried out,

and then the results were verified by the geological mapping collected from the two boxes,

known as the Kampung Pandan Roundabout and the Sungai Besi sites. Interestingly, the

results obtained from the two analyses have yielded the same ranges of domination, as in

the geological map analysis, the karst is predicted to form in the N030° – N070° and

N120° – N150° directions. In the geological structures mapping taken from the two


selected sites in Kuala Lumpur, a highly weathered and strongly deformed rock mass, thus

containing numerous karst features has also seen to develop in the N030° and N330°

directions; which is in a good agreement with the first predicted orientations. This in-direct

method of karst prediction orientations is proposed to be an alternative method to study the

subsurface orientations of karst adding to the currently used method of dye tracing testing

and other geophysical surveys. The joint/stream correspondence was found in the fracture

pattern developed in the limestone formation with the system of rivers meandering in the

non-carbonate rock formations. Joint orientation is important to the development of all

these stream orientations and presumably leads to the finding of subsurface karst in

carbonate formation. Nonetheless, the validation of karst prediction has just been made

based on a small part of the whole Kuala Lumpur area and hence much representative data

mapping and analysis are required to predict the formation of karst as a whole. Some

variations on the quality of rock mass properties can be seen between the Kampung Pandan

Roundabout and the Sungai Besi sites, considering the dominant structural events that

occurred at the vicinity of the areas. The Kampung Pandan Roundabout site is more

affected by the Ampang fault system and in the Sungai Besi site the limestone is

recrystallised by the close contact of a granitic body.

Acknowledgements

This research project is funded by a grant from the Service Department of Malaysia and

University Science of Malaysia.

References

[1] Bieniawski, Z. T., Engineering Rock Mechanic Classifications, John Wiley, New York,

1989.

[2] Chan, S. F., Hong, L. P., “Pile Foundations in Limestone Areas on Malaysia”, Proc.

8th Southeast Asian Geotechnical Conference, Kuala Lumpur, 11-15 March, 1985, 1,

P4/17-P4/28. 1985.

[3] Ericson, K., Migon, P. & Olvmo, “M., Fractures and drainage in the Granite

Mountainous Area, A study from Sierra Nevada, USA”, Journal Geomorphology, pp.

1-20, 2004.

[4] Gobbett, D. J., “The Lower Paleozoic Rocks of Kuala Lumpur, Malaysia”, Federation

Museum’s Journal, N. S. 9: pp. 67-79, 1964.

[5] Gobbett, D. J., “The Formation of Limestone Caves in Malaya”, Malay National

Journal. Vol. 19, pp. 4 – 12 pp, 1965.

[6] Gue, S. S. and Tan, Y. C., The Foundation System for Berjaya Times Square. Seminar

on Mega projects: A Study on Civil Engineering Projects, University Technology

Malaysia, Johor Bharu, Malaysia, 2001.

[7] Hatcher, R. D., Jr., Structural Geology: principles, concepts, and problems. 525.

Prentince-Hal, Inc, 1995.

[8] Herold, T., Jordan, P. & Zwahlen, F., “The influence of tectonic structures on karst

flow patterns in karstified limestone and aquitards in the Jura Mountains, Switzerland”,

Eclogae geol. Helv. 93, pp. 349-362, 2000.

[9]. Ibrahim, O. B and Fang, H. T., “Piled Foundation in Limestone Formation”, Proc 8th

.

Southeast Asian Geotechnical Conference, Kuala Lumpur, 11-15 March, 1985, 1.

Kuala Lumpur: Organising Committee, 1985, P4/45 -P4/52, 1985.

[10]Mayer, L., Menichetti, M., Nesci, O. & Savelli, D., “Morphotectonic approach to the

drainage analysis in the North Marche region, central Italy”, Quarterly International,

101-102, pp. 157-167, 2003.


[11]Mitchell, J. M., Foundations for the Pan Pacific Hotel on Pinnacled and Cavernous

Limestone. Proc. 8th

. Southeast Asian Geotechnical Conference, Kuala Lumpur, 11-15

March, 1985, 1. Kuala Lumpur: Organising Committee, 1985, P4/29 - P4/44, 1988.

[12]Ong, C. K., Petrology and structural geology of the Gombak Granite, Selangor.

Unpublished BSc (Hon) Thesis, University of Malaya, Malaysia, 1978.

[13]Paton, J. R., “The Origin of the Limestone Hills of Malaya”, Journal Tropical

Geography, Vol. 18, pp. 134 – 147, 1964.

[14]Pollalis, S., “The Petronas Twin Towers, Kuala Lumpur, Malaysia”, International

Cooperation and Information Transfer in the Realization of the World’s Tallest

Buildings. Harvard School Design. Publishing Division, Harvard Design School, pp. 1

-33, 2002.

[15]Price, N. J., Fault and joint development in brittle and semi-brittle rock. 176 p.

Pergamon Press, 1966.

[16]Shu, Y. K., “Some NW Trending faults in the Kuala Lumpur and other areas”, News.

Geol. Society Malaysia. 17, pp. 1 -15, 1969.

[17]SMART, “SMART Stormwater Management and Road Tunnel, Kuala Lumpur,

Malaysia”, Tunnels and Tunnelling International Supplement, May 2005, 2005.

[18]Sower, G. F., “Failures in Limestones in Humid Subtropics”, Proceedings American

Society Civil Engineers. 101 (GT8), pp. 771 – 787. 1976.

[19]Stauffer, P. H., “The Kuala Lumpur fault zone: A proposed major strike slip fault

across Malaya”, News. Geol. Society. Vol. 15, pp. 2 – 4. 1968.

[20]Tan, G. Y, Ting, W. H. & Toh, C. T., Foundation for Sri Mara Complex, Kuala

Lumpur. Proc. 8th

. Southeast Asian Geotechnical Conference, Kuala Lumpur, 11-15

March, 1985, 1. Kuala Lumpur: Organising Committee, 1985, pp. P4/68 -P4/72. 1985.

[21]Tan, B. K., Geological and Geotechnical Problems of Urban Centres in Malaysia. Proc

LANDPLAN II Symp., Kuala Lumpur, April 1984, Association of Geoscientists for

International Development, Kuala Lumpur, pp. 10-14. 1986a

[22]Tan, B. K., Use of Standard Penetration Test of Ground Interpretation – Some

Examples from Peninsular Malaysia. Proc. 5th

. International Congress, International

Association Engineering Geology, Buenos Aires, The International Association of

Engineering Geology, 839 – 846 pp, 1986b.

[23]Tan, B. K., and Komoo, I., Urban Geology: Case Study of Kuala Lumpur, Malaysia.

Engineering Geology, 28, pp. 71 -194, 1990.

[24]Tan, B. K., Project KLCC: Geology, soils and foundations. Warta Geology, 22 (2), pp.

73-74. 1996.

[25]Tjia, H. D., Strike slip faults in Malaysia. 24th

. International Geology Congress. 3, pp.

255-262. 1969.

[26]Verstappen, H. Th., Some Observations on Karst Development in the Malay

Archipelago. Journal Tropical Geography. Vol. 14, pp. 1 -10. 1960.

[27]Waltham, A. C. & Fookes, P. G., Engineering classification of karst ground conditions.

Quarterly Journal of Engineering Geology and Hydrogeology, 36, pp. 101-118. 2003.

[28]Yin, E. H., Geological map of Geology and Mineral Resources of Kuala Lumpur,

Malaysia. Geological Survey department Published. 1967.

[29]Zabidi, H & deFreitas, M. H., Re-evaluation of rock core logging for the prediction of

preferred orientations of karst in the Kuala Lumpur Limestone Formation, Eng Geol.

doi:10.1016/j.enggeo.2010.10.006. 2010.


preferred direction of karst in the kuala ... direction of karst in the kuala lumpur limestone...

Documents